From: Bill Davidsen (davidsen@tmr.com)
Date: Mon Aug 16 2004 - 13:48:42 CDT
Forrest J. Cavalier III wrote:
> Bill Davidsen wrote in part:
> > I think that we need to define followup because it will be
> > really dificult to have a cogent section on References and
> > In-Reply-To without that term.
>
> Any objections to this....?
>
> A "followup" is a message which is related, supplementary, or posted
> to respond
> to another message. A message includes a References header if and only
> if it is a followup.
The reason I defined a followup as a message with a references header is
that it's unambiguous. It covers all cases; it is not subject to
definitions of intention, relations, etc. You are starting with a
definition of what should have a references header, and defining which
posts should and should not have a references header. Unfortunately
there is no way software can detect violations of this, a reply without
references, a references header on a totally unrelated post (yes,
generated by mistake but possible).
Being an untructing person I prefer to simply have an unambiguous test
to identify what is and isn't a followup. So the software can make a
positive identification, because computers are good at that, and humans
can read the standard and decide if an article should have a references
header. Humans are crappy at that, but we have nothing better.
--
-bill davidsen (davidsen@tmr.com)
"The secret to procrastination is to put things off until the
last possible moment - but no longer" -me